ValueError:检查输入时出错:预期lstm_1_input具有3个维，但数组的形状为(393613，50)

问题描述：

我在Keras中出错，找不到解决方案.我已经搜索了整个互联网，但仍然没有答案^^ 这是我的代码.

I have a error in Keras and I can't find the solution. I have searched the whole internet and I have still no answer ^^ Here is my Code.

model = Sequential()
model.add(LSTM(32, return_sequences=True, input_shape=X.values.shape))
model.add(LSTM(32, return_sequences=True))
model.add(LSTM(32))
model.add(Dense(10, activation="softmax"))
model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=['accuracy'])

错误是第二行.它说:"ValueError:检查输入时出错:预期lstm_1_input具有3个维，但数组的形状为(393613，50)" 我的数据框X的形状正确. 当我尝试训练模型时，会弹出错误消息

The error is the second line. It says "ValueError: Error when checking input: expected lstm_1_input to have 3 dimensions, but got array with shape (393613, 50)" The Shape of my Dataframe X is correct. And when I try to train the model the Error pops up

model.fit(X.values, Y.values, batch_size=200, epochs=10, validation_split=0.05)

我希望有人可以帮助我:-)

I hope someone could help me :-)

顺便提一句.这是model.summary()

Btw. here is the model.summary()

lstm_1(LSTM)(无，393613，32)10624

lstm_1 (LSTM) (None, 393613, 32) 10624

lstm_2(LSTM)(无，393613，32)8320

lstm_2 (LSTM) (None, 393613, 32) 8320

lstm_3(LSTM)(无，32)8320

lstm_3 (LSTM) (None, 32) 8320

总参数:27,594 可训练的参数:27,594 不可训练的参数:0

Total params: 27,594 Trainable params: 27,594 Non-trainable params: 0

亲切的问候尼古拉斯.

答

在初始化第一层时，您正在传递2个值作为input_shape = X.values.shape

While initialising the first layer you are passing 2 values as input_shape =X.values.shape

keras已经希望每批的行数为NONE.在运行时，此值由batch_size=(在您的情况下为200)

keras already expects number of rows per batch as NONE. At runtime this value is determined by batch_size= (200 in your case)

因此，基本上，它在内部将第1层的输入形状更改为 NO_OF_FEATURES, NO_OF_ROWS_IN_DATA_SET, NO_OF_ROWS_PER_BATCH

So basically it internally changed shape of input for layer 1 as NO_OF_FEATURES, NO_OF_ROWS_IN_DATA_SET, NO_OF_ROWS_PER_BATCH

要解决此问题，您要做的就是将1参数作为input_shape传递，该参数没有功能. Keras已经接受NONE作为占位符，表示每个批次中没有行.

To fix this all you have to do is pass 1 parameter as input_shape, which is no of features. Keras already accepts NONE as a placeholder for no of rows per batch.

所以input_shape=(X.values.shape[1],)应该可以解决问题.

So input_shape=(X.values.shape[1],) should do the trick.

model.add(LSTM(32, return_sequences=True, input_shape=(X.values.shape[1],)))

ValueError:检查输入时出错:预期lstm_1_input具有3个维，但数组的形状为(393613，50)

相关推荐